Geothermal power generation facility abnormality determination system, abnormality determination device, data acquisition device, abnormality determination program, and data acquisition program

Abstract

To provide an abnormality determination system, an abnormality determination device, a data acquisition device, an abnormality determination program, and a data acquisition program for geothermal generation facilities that can determine abnormality in consideration of a factor affecting the performance of geothermal generation facilitate.SOLUTION: An abnormality determination system comprises a learning model 23 which has done machine learning so that when factor data obtained by measuring a factor affecting the flow rate of a working medium circulating in constitution equipment of geothermal generation facilities is input, an estimated performance value of each piece of constitution equipment is output, and acquires an estimated performance value by acquiring factor data and actual performance data, inputting factor data acquired from data acquisition means to the learning model 23, and performing arithmetic processing of the learning model 23, and determines that abnormality occurs to constitution equipment if the actual performance data deviates from the estimated performance value by a predetermined value or larger.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an abnormality determination system, an abnormality determination device, a data acquisition device, an abnormality determination program, and a data acquisition program for geothermal power generation equipment. [Background technology]

[0002] In recent years, small-scale geothermal power generation facilities that utilize hot spring heat (for example, see Patent Document 1) have been increasing. In general, the utilization rate of geothermal power generation facilities varies depending on environmental factors such as temperature changes of geothermal heat, river water, seawater, etc., and air temperature. The utilization rate also varies due to breakdowns of the equipment that constitutes the geothermal power plant and operational problems.

[0003] In order to identify the cause of such fluctuations in utilization rate, it is conceivable to determine that an abnormality has occurred in a component device when the performance of the component device changes by a certain amount or more. However, since the performance of the component device changes not only due to the above-mentioned breakdowns and operational troubles but also due to changes in the environment, it is difficult to conclude that the change in performance of the component device is due to a breakdown or the like.

[0004] In addition, although geothermal power generation facilities are equipped with management equipment that displays the measured values ​​of the temperature, pressure, flow rate, etc. of the working fluid flowing through the components, in most cases these measured values ​​are not provided as data. This makes it difficult to understand the performance of the components and determine abnormalities based on the data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2018-17173 A Summary of the Invention [Problem to be solved by the invention]

[0006] In consideration of the above circumstances, the present invention aims to provide an abnormality determination system, an abnormality determination device, a data acquisition device, an abnormality determination program, and a data acquisition program for geothermal power generation equipment that can determine abnormalities while taking into account factors that affect the performance of the geothermal power generation equipment. [Means for solving the problem]

[0007] An aspect of the present invention for achieving the above-mentioned object is an abnormality judgment system for a geothermal power generation facility that generates electricity by driving a turbine with a working medium, comprising: a learning model that has been subjected to machine learning so as to output an estimated performance value that estimates the performance of each component of the geothermal power generation facility when factor data obtained by measuring factors that affect the performance of the respective components of the geothermal power generation facility is input; a data acquisition means that acquires actual performance data obtained by measuring and analyzing the factor data; an operating state estimation means that inputs the factor data acquired from the data acquisition means into the learning model and acquires the estimated performance value by performing calculation processing of the learning model; and an abnormality judgment means that judges that an abnormality has occurred in the component equipment if the actual performance data acquired by the data acquisition means deviates from the estimated performance value by more than a predetermined value.

[0008] Another aspect of the present invention for achieving the above-mentioned object is an abnormality determination device for a geothermal power generation facility that generates electricity by driving a turbine with a working medium, comprising: a learning model that has been subjected to machine learning so as to output an estimated performance value that estimates the performance of each component of the geothermal power generation facility when factor data obtained by measuring factors that affect the performance of the respective components of the geothermal power generation facility is input; an operating state estimation means that inputs the factor data into the learning model and obtains the estimated performance value by executing calculation processing of the learning model; and an abnormality determination means that determines that an abnormality has occurred in the component equipment if actual performance data obtained by measuring and analyzing the factor data deviates from the estimated performance value by more than a predetermined value.

[0009] Another aspect of the present invention for achieving the above-mentioned object is an abnormality judgment program for a geothermal power generation facility that generates electricity by driving a turbine with a working medium, characterized in that the abnormality judgment device stores a learning model that has been subjected to machine learning so that when factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility is input, an estimated performance value that estimates the performance of the component equipment is output, the abnormality judgment program functions as: an operating state estimation means that inputs the factor data to the learning model and obtains the estimated performance value by executing calculation processing of the learning model; and an abnormality judgment means that determines that an abnormality has occurred in the component equipment if actual performance data obtained by measuring and analyzing the factor data deviates from the estimated performance value by more than a predetermined value.

[0010] Another aspect of the present invention for achieving the above-mentioned object is a data acquisition device capable of communicating with an abnormality judgment device for a geothermal power generation facility that generates electricity by driving a turbine with a working medium, and comprising a data acquisition means for acquiring actual performance data obtained by analyzing factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility, wherein the abnormality judgment device comprises a learning model that has been subjected to machine learning so as to output an estimated performance value that estimates the performance of the component equipment when the factor data is input, an operating state estimation means that inputs the factor data acquired from the data acquisition device into the learning model and acquires the estimated performance value by executing calculation processing of the learning model, and an abnormality judgment means that judges that an abnormality has occurred in the component equipment if the actual performance data acquired by the data acquisition device deviates from the estimated performance value by more than a predetermined value.

[0011] Another aspect of the present invention for achieving the above-mentioned object is a data acquisition program for geothermal power generation equipment, which comprises a data acquisition device capable of communicating with an abnormality judgment device for a geothermal power generation equipment that generates electricity by driving a turbine with a working medium, functioning as a data acquisition means for acquiring actual performance data obtained by analyzing factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation equipment, and the abnormality judgment device comprises a learning model that has been subjected to machine learning so as to output an estimated performance value that estimates the performance of the component equipment when factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation equipment is input, an operating state estimation means that inputs the factor data acquired from the data acquisition device into the learning model and acquires the estimated performance value by executing calculation processing of the learning model, and an abnormality judgment means that determines that an abnormality has occurred in the component equipment if the actual performance data acquired by the data acquisition device deviates from the estimated performance value by more than a predetermined value. Effect of the Invention

[0012] According to the present invention, it is an object of the present invention to provide an abnormality judgment system, an abnormality judgment device, a data acquisition device, an abnormality judgment program, and a data acquisition program for geothermal power generation equipment that can judge abnormalities while taking into account factors that affect the performance of the geothermal power generation equipment. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a configuration diagram of an abnormality determination system. [Diagram 2] FIG. 2 is a diagram illustrating a process flow of the abnormality determination system. [Diagram 3] FIG. 13 is a diagram showing time series changes in estimated flow rate and actual flow rate data. [Figure 4] FIG. 11 is a diagram showing time series changes in actual flow rate data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described. Note that the description of the embodiment is merely an example, and the present invention is not limited to the following description.

[0015] As shown in Figure 1, the abnormality judgment system 1 of this embodiment performs abnormality judgment of a geothermal power generation equipment 100, and is equipped with a data acquisition device 10 and an abnormality judgment device 20, and the data acquisition device 10 and the abnormality judgment device 20 are capable of communicating with each other via a communication means such as the Internet.

[0016] The geothermal power generation facility 100 is a so-called binary type geothermal power generation facility equipped with an evaporator, turbine, condenser, pump, generator, various measuring instruments, etc., although not shown in the figure, and includes a hot water system that produces hot water from geothermal heat, and a cooling system that produces cooling water from outside air, river water, seawater, etc. The working medium becomes steam due to geothermal heat, and the steam drives a turbine to generate electricity using a generator. The working medium discharged from the turbine is condensed in a condenser, and is again turned into steam in an evaporator. The pump is for pumping the working medium.

[0017] Such a geothermal power generation facility 100 has a monitor 101 that displays the operating status of various measuring instruments and components such as an evaporator. The measuring instruments measure the performance values ​​of the components of the geothermal power generation facility 100, and in this embodiment, for example, measure the flow rate of the working medium that has flowed through the components. Other measuring instruments are devices that measure environmental state quantities such as the air temperature around the components.

[0018] Specific examples of measuring instruments include instruments that measure the temperature, pressure, and flow rate of the working medium at the outlet of each component device, the temperature of the hot water obtained from the geothermal well, the temperature of the seawater used in the condenser, and the air temperature around the geothermal power generation facility 100.

[0019] The monitor 101 is a device that only displays the measured values ​​obtained by various measuring instruments. In the example of Fig. 1, the monitor 101 displays specific measured values ​​together with headings such as the hot water temperature, the seawater temperature, and the flow rate of the working medium of the evaporator (abbreviated as evaporator flow rate in the figure).

[0020] The data acquisition device 10 transmits the measurement values ​​obtained by various measuring instruments as factor data to the anomaly determination device. Specifically, although not shown, the data acquisition device 10 is a computer equipped with a CPU, RAM, a storage device such as a flash disk, an input / output device, a communication means, a camera, etc.

[0021] A data acquisition program is installed in the storage device of the data acquisition device 10, and the data acquisition program is loaded into the RAM and executed by the CPU. This data acquisition program causes the data acquisition device 10 to function as the data acquisition unit 11.

[0022] The data acquisition means acquires actual performance data obtained by measuring and analyzing the factor data. In this embodiment, the data acquisition means is composed of a data acquisition unit 11 that acquires the factor data, and a data analysis unit 21. The data acquisition unit 11 is provided in the data acquisition device 10, and the data analysis unit 21 is provided in the abnormality determination device 20.

[0023] The factor data is data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility 100. Examples of the factor data include data that indicates the temperature, pressure, flow rate, and composition of the working medium that flows through each component of the geothermal power generation facility 100. Other examples of the factor data include the temperature of the hot water obtained from the geothermal well, the temperature of the seawater used in the condenser, and the air temperature around the geothermal power generation facility 100.

[0024] The data acquiring unit 11 acquires the factor data as follows: In the case where a management device of the geothermal power generation facility 100 provides measurement values ​​obtained from each component device of the geothermal power generation facility via a known data communication means, the data acquiring unit 11 acquires the measurement data from the management device and uses it as factor data.

[0025] Furthermore, even if the geothermal power generation facility 100 does not provide measurement values ​​through a data communication means, it may provide a monitor 101 that displays measurement values ​​related to the operation status. In such a case, the measurement values ​​are obtained by image processing as follows. First, the monitor 101 is imaged by the camera 12 provided in the data acquisition device 10. The image captured of the monitor 101 displays measurement values ​​such as the temperature, pressure, flow rate, and composition of the working medium flowing through each component device, as well as the hot water temperature, and the data acquisition unit 11 converts these into numerical data through image analysis to obtain factor data.

[0026] As described above, the data acquisition unit 11 may employ a means for directly obtaining measured values ​​without image processing, or a means for indirectly obtaining measured values ​​without image processing, but only one of these may be employed, or both may be employed. For example, some measured values ​​may be obtained directly, and other measured values ​​may be obtained indirectly, depending on the situation in which the geothermal power generation facility 100 provides the measured values.

[0027] The timing at which the data acquisition unit 11 obtains measurement values ​​from the geothermal power generation equipment 100 via data communication means and converts them into factor data, and the timing at which the data acquisition unit 11 obtains factor data from the measurement values ​​displayed on the monitor 101, may be done at regular intervals, for example, every minute.

[0028] The data acquisition unit 11 may also execute a predetermined process on the factor data. For example, when some of the factor data obtained at each outlet of a plurality of component devices is missing, the data acquisition unit 11 may check whether or not data is missing by deleting all of the factor data obtained at that time. In addition, the data acquisition unit 11 may execute a process of averaging the factor data for a certain period of time.

[0029] The data acquiring unit 11 transmits the factor data acquired in this manner to the abnormality determination device 20 via a communication means. There is no limitation on the timing of data transmission, and the factor data may be transmitted collectively after a certain amount of data, for example, a day's worth, is accumulated, or the factor data may be transmitted each time it is acquired. The data acquiring unit 11 transmits the factor data in any format, such as CSV, TSV, JSON, or XML.

[0030] Furthermore, the data acquisition unit 11 may transmit the factor data directly to the abnormality determination device 20, or may store the factor data in a so-called cloud-type storage (storage device) provided via the Internet. In the latter case, the abnormality determination device 20 reads the factor data stored in the storage.

[0031] The abnormality determination device 20 performs an abnormality determination for the geothermal power generation facility 100 based on the factor data acquired from the data acquisition device 10. Specifically, the abnormality determination device 20 is a computer equipped with a CPU, RAM, storage devices such as a hard disk and a flash disk, input / output devices, communication means, etc., although not shown in the figure.

[0032] An abnormality determination program is installed in the storage device of the abnormality determination device 20, and is read into the RAM and executed by the CPU. This abnormality determination program causes the abnormality determination device 20 to function as a data analysis unit 21, a learning model formation means 22, an operating state estimation means 24, and an abnormality determination means 25.

[0033] The data analysis unit 21 acquires actual performance data by analyzing the factor data acquired by the data acquisition unit 11. The actual performance data is data that represents the performance of the geothermal power generation facility 100 and the performance of its component devices. Examples of the actual performance data include the generating end output, the sending end output, the turbine performance, the performance of each heat exchanger, and the performance of each pump.

[0034] An example of an analytical method for obtaining actual performance data from factor data is to obtain the actual performance data by solving a plant analysis model using the factor data. The plant analysis model represents the relational equations that should be satisfied by the temperature, pressure, flow rate, composition, etc. of the working fluid flowing through each component device (these are obtained as factor data). The data analysis unit 21 calculates the state of the working fluid of each component device by solving the relational equations. For each component device, an equation is defined for calculating the performance value of the component device from the state of the working fluid, and the data analysis unit 21 calculates the performance value (actual performance data) of each component device from the equation and the state of the working fluid.

[0035] The above-mentioned analysis method can be performed using EnergyWin®, which is disclosed in Japanese Patent No. 5522684 and Japanese Patent No. 5618322.

[0036] Of the factor data acquired from the data acquisition unit 11, the data from the period during which the geothermal power generation facility 100 is operating normally is referred to as learning data. The learning data is acquired as follows. A period during which a person has determined that the geothermal power generation facility 100 is operating normally based on its operating conditions is appropriately selected, and the factor data acquired during that period is extracted and used as the learning data. The learning data is used to form the learning model 23.

[0037] On the other hand, among the factor data, the data that is unknown as to whether it was acquired during normal driving or abnormal driving is called test data. The test data is used by the data analysis unit 21 and the driving state estimation means 24.

[0038] The learning model forming means 22 uses such learning data to perform machine learning that outputs an estimated performance value that is an estimate of the performance value of each component device when factor data is input, thereby forming a learning model 23. As a specific example of machine learning, for example, a neural network or regression can be adopted. The learning model forming means 22 stores the learning model 23 in the storage device of the anomaly determination device 20.

[0039] The learning model 23 is made up of functions, parameters, etc. that, when factor data is input, output estimated performance values ​​of each component device that would be obtained if the geothermal power generation equipment 100 were operating normally.

[0040] The driving state estimation means 24 inputs the test data into the learning model 23 and executes the calculation process of the learning model 23 to obtain an estimated performance value.

[0041] The abnormality determination means 25 determines that an abnormality has occurred in the constituent equipment if the actual performance data acquired from the data acquisition unit 11 and the data analysis unit 21 deviates by a predetermined value or more from the estimated performance value obtained by the operating state estimation means 24. As described above, the estimated performance value obtained using the learning model 23 is a value that would be measured if the geothermal power generation facility 100 were operating normally. Therefore, if the actual performance data deviates from the estimated performance value, it is considered that there is a high possibility that an abnormality has occurred in the constituent equipment of the geothermal power generation facility 100.

[0042] The flow of processing performed by the abnormality determination system 1 will be described with reference to FIG. The measured values ​​obtained directly from the geothermal power generation facility 100 and the measured values ​​of each component device displayed on the monitor 101 are converted into factor data by the data acquisition device 10 (data acquisition unit 11).

[0043] Next, the learning model forming means 22 performs machine learning using the factor data obtained during normal operation as learning data to form a learning model 23.

[0044] Next, the data analysis unit 21 uses, as test data, factor data among the factor data, which is unknown as to whether it has been obtained in a normal driving state or an abnormal driving state, to acquire an estimated performance value from the learning model 23. Meanwhile, the driving state estimation means 24 also performs an analysis based on the factor data that has been set as test data to acquire actual performance data.

[0045] Next, the abnormality determination means 25 compares the estimated performance value with the actual performance data and determines whether or not there is a deviation of a predetermined value or more. Whether or not there is a deviation of a predetermined value or more is determined by appropriately setting a ratio to the estimated performance value, such as whether or not the actual performance data is +5% or more or -5% or less with respect to the estimated performance value.

[0046] Fig. 3 is a diagram showing time series changes in estimated performance values ​​and actual performance data. The horizontal axis represents time, and the vertical axis represents turbine performance. The solid line, which represents the estimated performance value, represents the change in the estimated value of turbine performance. The dotted line, which represents the actual performance data, represents the change in turbine performance obtained by analyzing factor data. The actual performance data from time T1 to time T2 is sample S1, and the actual performance data from time T3 to time T4 is sample S2.

[0047] The abnormality determination means 25 determines that an abnormality has occurred in the turbine because the sample S1 of the actual performance data at time Ta deviates from the estimated performance value by more than a predetermined value. On the other hand, the abnormality determination means 25 determines that an abnormality has occurred in the turbine because the samples S1 and S2 at times other than time Ta are within the predetermined value from the estimated performance value. Such abnormality determination is performed for each component device of the geothermal power generation facility 100.

[0048] The abnormality determination device 20 makes the results of the abnormality determination available to the manager of the geothermal power generation facility 100 or the like on a smartphone 30 (see FIG. 1) via a communication means such as the Internet. This allows the manager to check the results of the abnormality determination and make a decision to repair or replace a component device that is determined to have an abnormality. This allows the component device to be repaired or replaced early on, thereby improving the utilization rate of the geothermal power generation facility 100.

[0049] Generally, in geothermal power generation equipment 100, pumps and the like are controlled to maximize output even when the hot water temperature, sea water temperature, air temperature, etc. As a result, the flow rate of the working medium, which is the performance of each component device, fluctuates, and even if the geothermal power generation equipment 100 is normal, as shown by characteristics S1 and S2 in Figure 3, the flow rate may differ significantly due to fluctuations in the hot water temperature, etc.

[0050] In such a geothermal power generation facility 100, a case will be described where an abnormality is judged based only on the actual performance data, as shown in Fig. 4. Samples S1 and S2 in Fig. 4 are the same as those in Fig. 3. If a threshold value Th is set for the actual performance data and an abnormality is judged if the threshold value Th is exceeded, sample S2 will be judged to be abnormal because it exceeds threshold Th at time T3. Moreover, sample S1 will be judged not to be abnormal because it is below threshold Th at time Ta. In this way, there are cases where a judgment result that is completely opposite to that in Fig. 3 is obtained.

[0051] When determining whether a condition is normal or abnormal by setting a threshold value for changes in actual performance data as described above, factors that affect the flow rate of the working medium, such as the hot water temperature, are not taken into consideration. This means that, as shown in Figure 4, it may not be possible to accurately determine whether a condition is normal or abnormal.

[0052] However, in the abnormality judgment system 1 according to this embodiment, abnormality judgment is performed not only using actual performance data but also using estimated performance values ​​obtained from a learning model 23 formed based on factor data obtained when the geothermal power generation equipment 100 is operating normally.

[0053] Therefore, even if relatively large actual performance data is obtained at time T3 of sample S2 shown in Figure 3, this is due to the influence of factors such as the hot water temperature, and since the deviation from the estimated performance value is small, it can be determined that the geothermal power generation facility 100 is operating normally. Also, even if not so large actual performance data is obtained at time Ta of sample S1, it can be determined that an abnormality has occurred in the geothermal power generation facility 100 because the deviation is large from the estimated performance value that takes into account factors such as the hot water temperature.

[0054] In this way, the abnormality determination system 1 according to this embodiment determines an abnormality while taking into consideration factors of environmental changes that affect the performance of the geothermal power generation facility 100, and can therefore more accurately determine an abnormality in the geothermal power generation facility 100. As a result of accurately determining an abnormality in this way, for example, it is possible to find a faulty component device early and improve performance by repairing or replacing it, thereby improving the utilization rate of the geothermal power generation facility 100.

[0055] Furthermore, the data acquisition device 10 according to this embodiment acquires factor data from measurement values ​​displayed on a monitor 101 provided in the geothermal power generation facility 100. As a result, even if the geothermal power generation facility 100 does not provide measurement values ​​of various measuring instruments as data, it is possible to notify an abnormality in the geothermal power generation facility 100 at an early stage and provide useful information for taking measures such as repair or replacement.

[0056] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and addition, omission, substitution, and other modifications of the configuration are possible without departing from the spirit of the present invention.

[0057] For example, the abnormality determination device 20 does not necessarily have to include the learning model formation means 22. In this case, the learning model 23 may be formed by another information processing device, and the learning model 23 may be stored in the storage device of the abnormality determination device 20.

[0058] In addition, the configuration in which the data acquisition unit 11 is provided in the data acquisition device 10 and the data analysis unit 21 is provided in the abnormality determination device 20 has been exemplified, but is not limited to this. For example, both the data acquisition unit 11 and the data analysis unit 21 may be provided in the data acquisition device 10, or both may be provided in the abnormality determination device.

[0059] Furthermore, the data acquisition unit 11, the data analysis unit 21, the learning model formation means 22, the driving state estimation means 24, and the abnormality determination means 25 do not need to be provided in one abnormality determination device 20, and may be provided in multiple computers. For example, a computer equipped with the data analysis unit 21 and the learning model formation means 22 is caused to create a learning model 23. Then, the learning model 23 is stored in a computer equipped with the driving state estimation means 24 and the abnormality determination means 25. In this way, a series of processes as shown in FIG. 2 may be performed by distributing them as a whole to multiple computers.

[0060] Furthermore, there is no particular limit to the number of data acquisition devices 10 and anomaly determination devices 20. For example, multiple data acquisition devices 10 may be used depending on the number and locations of the monitors 101. Furthermore, multiple anomaly determination devices 20 may be used to distribute the load and improve calculation speed.

[0061] Also, the data acquisition device 10 is provided with a data acquisition unit to convert the measurement values ​​into factor data, but the present invention is not limited to this configuration. For example, the data acquisition device 10 may transmit an image captured by a camera to the abnormality determination device 20 via a communication means. The abnormality determination device 20 may convert the measurement values ​​into factor data based on the image. In this case, the camera and communication means provided in the data acquisition device 10 and the function of the abnormality determination device to analyze the image and convert it into factor data constitute the data acquisition unit.

[0062] In addition, the data acquisition device 10 and the abnormality determination device 20 do not need to be separate devices. In other words, a single information processing device may store the learning model and execute a program including a data acquisition means, a driving state estimation means, and an abnormality determination means, or a program including a learning model formation means. [Explanation of symbols]

[0063] 1...abnormality determination system, 10...data acquisition device, 11...data acquisition unit, 12...camera, 20...abnormality determination device, 22...learning model formation means, 23...learning model, 24...operating state estimation means, 25...abnormality determination means, 30...smartphone, 100...geothermal power generation facility, 101...monitor

Claims

1. An abnormality determination system for a geothermal power generation facility that generates power by driving a turbine with a working medium, comprising: a learning model that performs machine learning so as to output an estimated performance value that estimates the performance of each component of the geothermal power generation facility when factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility is input; a data acquisition unit configured to acquire the factor data and a data analysis unit to acquire actual performance data by solving a plant analysis model using the factor data acquired by the data acquisition unit; a driving state estimating means for inputting the factor data acquired from the data acquiring means into the learning model and acquiring the estimated performance value by executing a calculation process of the learning model; and an abnormality determination means for determining that an abnormality has occurred in the component device if the actual performance data acquired by the data acquisition means deviates from the estimated performance value by a predetermined value or more. The abnormality determination system for geothermal power generation equipment is characterized by the above.

2. The abnormality determination system for a geothermal power generation facility according to claim 1, A learning model forming means for forming the learning model by performing machine learning that outputs the estimated performance value when the factor data during a period during which the geothermal power generation facility is normally operating is inputted. The abnormality determination system for geothermal power generation equipment is characterized by the above.

3. The abnormality determination system for a geothermal power generation facility according to claim 1 or 2, the geothermal power generation facility includes a monitor that measures the factor data and displays the measurement value; The data acquisition unit acquires the factor data by converting the measurement values ​​displayed on the monitor into numerical values ​​through image analysis. The abnormality determination system for geothermal power generation equipment is characterized by the above.

4. An abnormality determination device for a geothermal power generation facility that generates electricity by driving a turbine with a working medium, a learning model that performs machine learning so as to output an estimated performance value that estimates the performance of each component of the geothermal power generation facility when factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility is input; a driving state estimating means for inputting the factor data into the learning model and executing a calculation process of the learning model to obtain the estimated performance value; a data analysis unit that obtains actual performance data by solving a plant analysis model using the factor data; and an abnormality determination means for determining that an abnormality has occurred in the component device if the actual performance data obtained by the data analysis unit deviates from the estimated performance value by a predetermined value or more. The abnormality determination device for geothermal power generation equipment is characterized in that

5. An abnormality determination program for a geothermal power generation facility that generates power by driving a turbine using a working medium, comprising: an anomaly determination device that stores a learning model that has been subjected to machine learning so that, when factor data obtained by measuring factors that affect the performance of each component of the geothermal power generation facility is input, an estimated performance value that estimates the performance of the component equipment is output; a driving state estimating means for inputting the factor data into the learning model and executing a calculation process of the learning model to obtain the estimated performance value; a data analysis unit that obtains actual performance data by solving a plant analysis model using the factor data; and functioning as an abnormality determination means for determining that an abnormality has occurred in the component device if actual performance data obtained by measuring and analyzing the factor data deviates from the estimated performance value by a predetermined value or more. A geothermal power generation facility abnormality determination program comprising:

6. A data acquisition device capable of communicating with the abnormality determination device according to claim 4, A data acquisition unit that acquires factor data by measuring factors that affect the performance of each component of the geothermal power generation facility. A data acquisition device for geothermal power generation equipment.

7. A data acquisition device capable of communicating with the abnormality determination device according to claim 4, The device functions as a data acquisition unit that acquires factor data by measuring factors that affect the performance of each component of the geothermal power generation facility. A data acquisition program for geothermal power generation equipment.

Images